We have shown that tumor imaging with a radiolabeled antibody (PA) against carcinoembryonic antigen (CEA) can be improved by rapidly removing excess PA from the blood with a second antibody (SA; 1,2). Although useful for imaging, this anti-antibody method may have greater utility for therapeutic applications. Thus, the principal aim of this project is to determine the efficacy of using a second antibody (SA) to improve radioimmunotherapy (RAIT) of cancer. Removal of nontumor targeted, blood-pool radioantibody reduces the toxicity of RAIT treatment by removing harmful amounts of radioactivity from the blood and nontumor tissues, while maintaining a high retention of radioantibody in the tumor. We will examine the extent of RAIT improvement using this SA method in nude mice bearing tumors subcutaneously or in visceral organs (liver and lungs), or intraperitoneally, to determine how tumors located in these various sites are influenced by radioantibody treatment. Comparisons will be made to single and multiple injection of the radioantibody alone with or without SA, with SA as intact IgG or F(ab') 2 fragments, and to F(ab') 2 fragments of radiolabeled PA. Both I-131-and Y-90-Labeled PA will be tested to compare the utility of the SA to reduce toxicity and improve tumoricidal effects with PA labeled with iodine or a metal. Additional studies will assess the ability of the SA to reduce the immune response against the PA by rapidly removing it from the blood. In this case mouse anti-goat IgG will be given to BALB/C mice at various times after their injection with goat IgG. Mouse anti-goat IgG will be used to provide a SA that is not immunogenic. Finally, we will evaluate the efficacy and possible toxic-side effects of the SA method in rabbits with a mouse monoclonal anti-CEA antibody and a baboon SA that could be used in future clinical trials. These studies will determine what role the SA may have in improving RAIT in man.